New API - Open Source AI Large Model Gateway and Asset Management System

What is the New API?

New API is a next-generation AI gateway and asset management system. As an AI foundation platform, it provides unified infrastructure access to over 30 mainstream AI services globally (OpenAI, Claude, Gemini, DeepSeek, etc.). Core platform features include a unified OpenAI-compatible interface, intelligent routing load balancing, granular billing and access control, and real-time data dashboards. The platform supports advanced functions such as multi-format conversion, inference intensity control, and cached billing. Adopting the AGPLv3 open-source license, it supports one-click Docker deployment and is suitable for individual developers and enterprise-level multi-tenant scenarios.

Main functions of the new API

• Unified Interface ManagementProvides a single API endpoint compatible with the OpenAI format, seamlessly connecting to 30+ mainstream AI service providers worldwide.
• Intelligent routing schedulingIt supports multi-channel load balancing, automatic failover, and weighted random distribution to ensure high service availability.
• Fine-grained billing systemSupports billing by number of times or by quantity, prepaid top-up, multi-rate configuration, and cached billing.
• Security Access ControlIt provides token group management, model access restrictions, API call auditing, and multi-platform authorization login.
• Format conversion capabilitySupports mutual conversion between various API formats such as OpenAI, Claude Messages, and Google Gemini.
• Reasoning strength controlIt supports flexibly setting different levels of reasoning intensity (high, medium, and low) by using the model name suffix.
• Real-time data dashboardIt provides data insights through a visual console, usage statistics analysis, and cost monitoring.

Key information and usage requirements of the new API

• Project PositioningNext-generation AI gateway and asset management system, AI foundation platform
• Open source licenseGNU AGPLv3 (free to use, SaaS deployment requires open source)
• Compatibility baseDeveloped based on the One API, fully compatible with the original database.
• Supported languagesSimplified Chinese, Traditional Chinese, English, French, Japanese
• Deployment methodDocker / Docker Compose / BT Panel
• databaseSQLite (default) / MySQL ≥ 5.7.8 / PostgreSQL ≥ 9.6
• Docker images:calciumion/new-api:latest

The core advantages of the new API

• Unified AccessThrough a single API endpoint compatible with the OpenAI format, it can seamlessly connect to 30+ mainstream AI service providers worldwide, completely eliminating the tedious work of connecting multiple platforms.
• Smart RouterThe platform has built-in multi-channel load balancing, automatic failover, and weighted random distribution mechanisms to ensure high availability and request stability of AI services.
• Cost optimizationIt supports caching billing, pay-as-you-go or pay-per-use billing, and flexible configuration of multiple rates, helping users achieve refined cost control and expense management.
• Format interoperabilityIt provides the ability to freely convert between multiple API formats such as OpenAI, Claude Messages, and Google Gemini, significantly reducing the access barrier for different models.
• Ready to use right out of the boxSupports one-click Docker deployment, is fully compatible with One API database, and provides visual installation via BT Panel, greatly simplifying the deployment process.

How to use the New API

• Deployment and installationAfter cloning the project repository to your local machine, edit the configuration file, execute the Docker command to start the service, and access the system via the default port 3000 in your browser.
• Initial configurationLog in to the management backend to set up an administrator account, add the API keys of each AI service provider in channel management, and configure weights and failover strategies.
• Create access credentialsCreate an API Key on the token management page, set the limit, validity period and available model range, and assign independent credentials for different scenarios to achieve permission isolation.
• Access and usePoint the application API base address to the New API deployment address, replace the original key with the generated token, and maintain the OpenAI standard format for seamless calling of multi-platform models.

Project address for New API

• Project official websitehttps://www.newapi.ai/
• GitHub repositoryhttps://github.com/QuantumNous/new-api

Comparison of similar products with the new API

Application scenarios of the new API

• Website building for individual developersQuickly build a private AI interface relay station, uniformly manage API keys from multiple platforms, control personal usage costs through precise billing, and avoid the cumbersome operation of frequently switching between different service providers.
• Startup team product developmentIt provides stable multi-model backend support for AI applications, ensuring high availability of product services, while monitoring usage and costs through data dashboards to optimize resource allocation.
• Enterprise internal AI platformThe platform supports the construction of an enterprise-level AI asset management system, which unifies the management of model access and expenses, meets compliance requirements, and improves management efficiency.
• AI Model Comparison TestUsers can quickly switch between different vendor models through a unified interface and compare the performance of GPT, Claude, Gemini, DeepSeek, etc. in real-world tasks to assist in technology selection decisions.